© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 3-RES-JB
    11. 23656
    Prev
    Next
    INDEX
    Explanations

    phrases related to customer interactions and services

    oai_token-act-pair · gpt-3.5-turbo

    references to customers and customer-related topics

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Alz
    -0.78
    olon
    -0.70
     Hat
    -0.68
     Soccer
    -0.67
     Osc
    -0.67
    apo
    -0.66
    FLAG
    -0.65
     Glob
    -0.65
     Mong
    -0.64
     IMAGES
    -0.64
    POSITIVE LOGITS
    ele
    0.99
     customers
    0.94
    umers
    0.85
    hip
    0.83
    atical
    0.82
     tuning
    0.80
    coni
    0.79
    untled
    0.75
     testers
    0.75
     refunds
    0.73
    Activations Density 0.023%

    No Known Activations